This proposal seeks funding to measure physical activity objectively using the Actigraph GT3X accelerometer, a device that can accurately measure movement accelerations in three planes, among 21,249 women in the Women's Health Study and to follow them over 4 years for mortality. Already available are data from annual questionnaires over 15 years on risk factors (including self-reported physical activity every 2-3 years) and medical history in women, as well as plasma biomarker and whole genome scan data from blood samples collected at baseline. Morbidity follow-up is well over 90% and mortality follow-up is virtually 100%. In 2008, the federal government released its first-ever Physical Activity Guidelines for Americans, recommending the types and amounts of physical activity believed to offer substantial health benefits. The basis for these guidelines comes almost exclusively from observational epidemiologic studies of clinical endpoints, with physical activity assessed by self-report. Supporting such studies are data from randomized controlled trials of supervised exercise and cardiovascular risk factors, providing plausible biologic underpinnings for the lower risks of chronic diseases with physical activity. However, national surveys using self-reported physical activity show 47-50% of American adults meet physical activity recommendations, but national surveys measuring physical activity objectively using accelerometers indicate that <5% do so. Additionally, recent data suggest that sedentary behavior (e.g., time sitting) may be an independent risk factor, separate from physical activity, for chronic diseases. Sedentary behavior is imprecisely measured using self- reports, but can be accurately determined using accelerometers. Thus, the present proposal seeks to clarify the relations between physical activity, as well as sedentary behavior, and mortality by objectively measuring physical activity and sedentary behavior. Women will be asked to wear the Actigraph accelerometer for 1 week. Women will be recruited on a rolling basis over 2 years. Data from the accelerometers will classify women according to the time spent in light, moderate, and vigorous intensity physical activity (or combinations thereof), and sedentary behavior. At the end of the study, we will use the National Death Index to ascertain decedents. We will have sufficient statistical power to determine associations between physical activity, or sedentary behavior, and all-cause mortality. We also will be able to compare self-reported and accelerometer-measured physical activity, providing useful data on measurement error correction for other epidemiologic studies which may not have the resources to objectively measure physical activity. Finally, this study will be of tremendous advantage for future grant applications to cost- efficiently address the associations of objectively measured physical activity with other clinical endpoints which will accumulate in sufficient numbers over time, such as incident cardiovascular disease, cancer, and type 2 diabetes, all major causes of morbidity and mortality, as well as to examine physical activity-gene interactions. 1